user/sven/linux.git/fs/ext4, branch v6.18.1

ext4: add i_data_sem protection in ext4_destroy_inline_data_nolock()

2025-12-12T17:42:45Z

commit 0cd8feea8777f8d9b9a862b89c688b049a5c8475 upstream. Fix a race between inline data destruction and block mapping. The function ext4_destroy_inline_data_nolock() changes the inode data layout by clearing EXT4_INODE_INLINE_DATA and setting EXT4_INODE_EXTENTS. At the same time, another thread may execute ext4_map_blocks(), which tests EXT4_INODE_EXTENTS to decide whether to call ext4_ext_map_blocks() or ext4_ind_map_blocks(). Without i_data_sem protection, ext4_ind_map_blocks() may receive inode with EXT4_INODE_EXTENTS flag and triggering assert. kernel BUG at fs/ext4/indirect.c:546! EXT4-fs (loop2): unmounting filesystem. invalid opcode: 0000 [#1] PREEMPT SMP KASAN NOPTI Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 RIP: 0010:ext4_ind_map_blocks.cold+0x2b/0x5a fs/ext4/indirect.c:546 Call Trace: ext4_map_blocks+0xb9b/0x16f0 fs/ext4/inode.c:681 _ext4_get_block+0x242/0x590 fs/ext4/inode.c:822 ext4_block_write_begin+0x48b/0x12c0 fs/ext4/inode.c:1124 ext4_write_begin+0x598/0xef0 fs/ext4/inode.c:1255 ext4_da_write_begin+0x21e/0x9c0 fs/ext4/inode.c:3000 generic_perform_write+0x259/0x5d0 mm/filemap.c:3846 ext4_buffered_write_iter+0x15b/0x470 fs/ext4/file.c:285 ext4_file_write_iter+0x8e0/0x17f0 fs/ext4/file.c:679 call_write_iter include/linux/fs.h:2271 [inline] do_iter_readv_writev+0x212/0x3c0 fs/read_write.c:735 do_iter_write+0x186/0x710 fs/read_write.c:861 vfs_iter_write+0x70/0xa0 fs/read_write.c:902 iter_file_splice_write+0x73b/0xc90 fs/splice.c:685 do_splice_from fs/splice.c:763 [inline] direct_splice_actor+0x10f/0x170 fs/splice.c:950 splice_direct_to_actor+0x33a/0xa10 fs/splice.c:896 do_splice_direct+0x1a9/0x280 fs/splice.c:1002 do_sendfile+0xb13/0x12c0 fs/read_write.c:1255 __do_sys_sendfile64 fs/read_write.c:1323 [inline] __se_sys_sendfile64 fs/read_write.c:1309 [inline] __x64_sys_sendfile64+0x1cf/0x210 fs/read_write.c:1309 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x35/0x80 arch/x86/entry/common.c:81 entry_SYSCALL_64_after_hwframe+0x6e/0xd8 Fixes: c755e251357a ("ext4: fix deadlock between inline_data and ext4_expand_extra_isize_ea()") Cc: stable@vger.kernel.org # v4.11+ Signed-off-by: Alexey Nepomnyashih Message-ID: <20251104093326.697381-1-sdl@nppct.ru> Signed-off-by: Theodore Ts'o Signed-off-by: Greg Kroah-Hartman

ext4: refresh inline data size before write operations

2025-12-12T17:42:44Z

commit 892e1cf17555735e9d021ab036c36bc7b58b0e3b upstream. The cached ei->i_inline_size can become stale between the initial size check and when ext4_update_inline_data()/ext4_create_inline_data() use it. Although ext4_get_max_inline_size() reads the correct value at the time of the check, concurrent xattr operations can modify i_inline_size before ext4_write_lock_xattr() is acquired. This causes ext4_update_inline_data() and ext4_create_inline_data() to work with stale capacity values, leading to a BUG_ON() crash in ext4_write_inline_data(): kernel BUG at fs/ext4/inline.c:1331! BUG_ON(pos + len > EXT4_I(inode)->i_inline_size); The race window: 1. ext4_get_max_inline_size() reads i_inline_size = 60 (correct) 2. Size check passes for 50-byte write 3. [Another thread adds xattr, i_inline_size changes to 40] 4. ext4_write_lock_xattr() acquires lock 5. ext4_update_inline_data() uses stale i_inline_size = 60 6. Attempts to write 50 bytes but only 40 bytes actually available 7. BUG_ON() triggers Fix this by recalculating i_inline_size via ext4_find_inline_data_nolock() immediately after acquiring xattr_sem. This ensures ext4_update_inline_data() and ext4_create_inline_data() work with current values that are protected from concurrent modifications. This is similar to commit a54c4613dac1 ("ext4: fix race writing to an inline_data file while its xattrs are changing") which fixed i_inline_off staleness. This patch addresses the related i_inline_size staleness issue. Reported-by: syzbot+f3185be57d7e8dda32b8@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?extid=f3185be57d7e8dda32b8 Cc: stable@kernel.org Signed-off-by: Deepanshu Kartikey Message-ID: <20251020060936.474314-1-kartikey406@gmail.com> Signed-off-by: Theodore Ts'o Signed-off-by: Greg Kroah-Hartman

Merge tag 'ext4_for_linus-6.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

2025-10-15T14:51:57Z

Pull ext4 bug fixes from Ted Ts'o: - Fix regression caused by removing CONFIG_EXT3_FS when testing some very old defconfigs - Avoid a BUG_ON when opening a file on a maliciously corrupted file system - Avoid mm warnings when freeing a very large orphan file metadata - Avoid a theoretical races between metadata writeback and checkpoints (it's very hard to hit in practice, since the race requires that the writeback take a very long time) * tag 'ext4_for_linus-6.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: Use CONFIG_EXT4_FS instead of CONFIG_EXT3_FS in all of the defconfigs ext4: free orphan info with kvfree ext4: detect invalid INLINE_DATA + EXTENTS flag combination ext4, doc: fix and improve directory hash tree description ext4: wait for ongoing I/O to complete before freeing blocks jbd2: ensure that all ongoing I/O complete before freeing blocks

ext4: free orphan info with kvfree

2025-10-10T20:42:51Z

Orphan info is now getting allocated with kvmalloc_array(). Free it with kvfree() instead of kfree() to avoid complaints from mm. Reported-by: Chris Mason Fixes: 0a6ce20c1564 ("ext4: verify orphan file size is not too big") Cc: stable@vger.kernel.org Signed-off-by: Jan Kara Message-ID: <20251007134936.7291-2-jack@suse.cz> Signed-off-by: Theodore Ts'o

ext4: detect invalid INLINE_DATA + EXTENTS flag combination

2025-10-10T20:42:51Z

syzbot reported a BUG_ON in ext4_es_cache_extent() when opening a verity file on a corrupted ext4 filesystem mounted without a journal. The issue is that the filesystem has an inode with both the INLINE_DATA and EXTENTS flags set: EXT4-fs error (device loop0): ext4_cache_extents:545: inode #15: comm syz.0.17: corrupted extent tree: lblk 0 < prev 66 Investigation revealed that the inode has both flags set: DEBUG: inode 15 - flag=1, i_inline_off=164, has_inline=1, extents_flag=1 This is an invalid combination since an inode should have either: - INLINE_DATA: data stored directly in the inode - EXTENTS: data stored in extent-mapped blocks Having both flags causes ext4_has_inline_data() to return true, skipping extent tree validation in __ext4_iget(). The unvalidated out-of-order extents then trigger a BUG_ON in ext4_es_cache_extent() due to integer underflow when calculating hole sizes. Fix this by detecting this invalid flag combination early in ext4_iget() and rejecting the corrupted inode. Cc: stable@kernel.org Reported-and-tested-by: syzbot+038b7bf43423e132b308@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=038b7bf43423e132b308 Suggested-by: Zhang Yi Signed-off-by: Deepanshu Kartikey Reviewed-by: Zhang Yi Message-ID: <20250930112810.315095-1-kartikey406@gmail.com> Signed-off-by: Theodore Ts'o

ext4: wait for ongoing I/O to complete before freeing blocks

2025-10-10T17:34:35Z

When freeing metadata blocks in nojournal mode, ext4_forget() calls bforget() to clear the dirty flag on the buffer_head and remvoe associated mappings. This is acceptable if the metadata has not yet begun to be written back. However, if the write-back has already started but is not yet completed, ext4_forget() will have no effect. Subsequently, ext4_mb_clear_bb() will immediately return the block to the mb allocator. This block can then be reallocated immediately, potentially causing an data corruption issue. Fix this by clearing the buffer's dirty flag and waiting for the ongoing I/O to complete, ensuring that no further writes to stale data will occur. Fixes: 16e08b14a455 ("ext4: cleanup clean_bdev_aliases() calls") Cc: stable@kernel.org Reported-by: Gao Xiang Closes: https://lore.kernel.org/linux-ext4/a9417096-9549-4441-9878-b1955b899b4e@huaweicloud.com/ Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Message-ID: <20250916093337.3161016-3-yi.zhang@huaweicloud.com> Signed-off-by: Theodore Ts'o

Merge tag 'ext4_for_linus-6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

2025-10-03T20:47:10Z

Pull ext4 updates from Ted Ts'o: "New ext4 features: - Add support so tune2fs can modify/update the superblock using an ioctl, without needing write access to the block device - Add support for 32-bit reserved uid's and gid's Bug fixes: - Fix potential warnings and other failures caused by corrupted / fuzzed file systems - Fail unaligned direct I/O write with EINVAL instead of silently falling back to buffered I/O - Correectly handle fsmap queries for metadata mappings - Avoid journal stalls caused by writeback throttling - Add some missing GFP_NOFAIL flags to avoid potential deadlocks under extremem memory pressure Cleanups: - Remove obsolete EXT3 Kconfigs" * tag 'ext4_for_linus-6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: ext4: fix checks for orphan inodes ext4: validate ea_ino and size in check_xattrs ext4: guard against EA inode refcount underflow in xattr update ext4: implemet new ioctls to set and get superblock parameters ext4: add support for 32-bit default reserved uid and gid values ext4: avoid potential buffer over-read in parse_apply_sb_mount_options() ext4: fix an off-by-one issue during moving extents ext4: increase i_disksize to offset + len in ext4_update_disksize_before_punch() ext4: verify orphan file size is not too big ext4: fail unaligned direct IO write with EINVAL ext4: correctly handle queries for metadata mappings ext4: increase IO priority of fastcommit ext4: remove obsolete EXT3 config options jbd2: increase IO priority of checkpoint ext4: fix potential null deref in ext4_mb_init() ext4: add ext4_sb_bread_nofail() helper function for ext4_free_branches() ext4: replace min/max nesting with clamp() fs: ext4: change GFP_KERNEL to GFP_NOFS to avoid deadlock

Merge tag 'vfs-6.18-rc1.workqueue' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs

2025-09-29T17:27:17Z

Pull vfs workqueue updates from Christian Brauner: "This contains various workqueue changes affecting the filesystem layer. Currently if a user enqueue a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This replaces the use of system_wq and system_unbound_wq. system_wq is a per-CPU workqueue which isn't very obvious from the name and system_unbound_wq is to be used when locality is not required. So this renames system_wq to system_percpu_wq, and system_unbound_wq to system_dfl_wq. This also adds a new WQ_PERCPU flag to allow the fs subsystem users to explicitly request the use of per-CPU behavior. Both WQ_UNBOUND and WQ_PERCPU flags coexist for one release cycle to allow callers to transition their calls. WQ_UNBOUND will be removed in a next release cycle" * tag 'vfs-6.18-rc1.workqueue' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: fs: WQ_PERCPU added to alloc_workqueue users fs: replace use of system_wq with system_percpu_wq fs: replace use of system_unbound_wq with system_dfl_wq

Merge tag 'vfs-6.18-rc1.inode' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs

2025-09-29T16:42:30Z

Pull vfs inode updates from Christian Brauner: "This contains a series I originally wrote and that Eric brought over the finish line. It moves out the i_crypt_info and i_verity_info pointers out of 'struct inode' and into the fs-specific part of the inode. So now the few filesytems that actually make use of this pay the price in their own private inode storage instead of forcing it upon every user of struct inode. The pointer for the crypt and verity info is simply found by storing an offset to its address in struct fsverity_operations and struct fscrypt_operations. This shrinks struct inode by 16 bytes. I hope to move a lot more out of it in the future so that struct inode becomes really just about very core stuff that we need, much like struct dentry and struct file, instead of the dumping ground it has become over the years. On top of this are a various changes associated with the ongoing inode lifetime handling rework that multiple people are pushing forward: - Stop accessing inode->i_count directly in f2fs and gfs2. They simply should use the __iget() and iput() helpers - Make the i_state flags an enum - Rework the iput() logic Currently, if we are the last iput, and we have the I_DIRTY_TIME bit set, we will grab a reference on the inode again and then mark it dirty and then redo the put. This is to make sure we delay the time update for as long as possible We can rework this logic to simply dec i_count if it is not 1, and if it is do the time update while still holding the i_count reference Then we can replace the atomic_dec_and_lock with locking the ->i_lock and doing atomic_dec_and_test, since we did the atomic_add_unless above - Add an icount_read() helper and convert everyone that accesses inode->i_count directly for this purpose to use the helper - Expand dump_inode() to dump more information about an inode helping in debugging - Add some might_sleep() annotations to iput() and associated helpers" * tag 'vfs-6.18-rc1.inode' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: fs: add might_sleep() annotation to iput() and more fs: expand dump_inode() inode: fix whitespace issues fs: add an icount_read helper fs: rework iput logic fs: make the i_state flags an enum fs: stop accessing ->i_count directly in f2fs and gfs2 fsverity: check IS_VERITY() in fsverity_cleanup_inode() fs: remove inode::i_verity_info btrfs: move verity info pointer to fs-specific part of inode f2fs: move verity info pointer to fs-specific part of inode ext4: move verity info pointer to fs-specific part of inode fsverity: add support for info in fs-specific part of inode fs: remove inode::i_crypt_info ceph: move crypt info pointer to fs-specific part of inode ubifs: move crypt info pointer to fs-specific part of inode f2fs: move crypt info pointer to fs-specific part of inode ext4: move crypt info pointer to fs-specific part of inode fscrypt: add support for info in fs-specific part of inode fscrypt: replace raw loads of info pointer with helper function

Merge tag 'vfs-6.18-rc1.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs

2025-09-29T16:03:07Z

Pull misc vfs updates from Christian Brauner: "This contains the usual selections of misc updates for this cycle. Features: - Add "initramfs_options" parameter to set initramfs mount options. This allows to add specific mount options to the rootfs to e.g., limit the memory size - Add RWF_NOSIGNAL flag for pwritev2() Add RWF_NOSIGNAL flag for pwritev2. This flag prevents the SIGPIPE signal from being raised when writing on disconnected pipes or sockets. The flag is handled directly by the pipe filesystem and converted to the existing MSG_NOSIGNAL flag for sockets - Allow to pass pid namespace as procfs mount option Ever since the introduction of pid namespaces, procfs has had very implicit behaviour surrounding them (the pidns used by a procfs mount is auto-selected based on the mounting process's active pidns, and the pidns itself is basically hidden once the mount has been constructed) This implicit behaviour has historically meant that userspace was required to do some special dances in order to configure the pidns of a procfs mount as desired. Examples include: * In order to bypass the mnt_too_revealing() check, Kubernetes creates a procfs mount from an empty pidns so that user namespaced containers can be nested (without this, the nested containers would fail to mount procfs) But this requires forking off a helper process because you cannot just one-shot this using mount(2) * Container runtimes in general need to fork into a container before configuring its mounts, which can lead to security issues in the case of shared-pidns containers (a privileged process in the pidns can interact with your container runtime process) While SUID_DUMP_DISABLE and user namespaces make this less of an issue, the strict need for this due to a minor uAPI wart is kind of unfortunate Things would be much easier if there was a way for userspace to just specify the pidns they want. So this pull request contains changes to implement a new "pidns" argument which can be set using fsconfig(2): fsconfig(procfd, FSCONFIG_SET_FD, "pidns", NULL, nsfd); fsconfig(procfd, FSCONFIG_SET_STRING, "pidns", "/proc/self/ns/pid", 0); or classic mount(2) / mount(8): // mount -t proc -o pidns=/proc/self/ns/pid proc /tmp/proc mount("proc", "/tmp/proc", "proc", MS_..., "pidns=/proc/self/ns/pid"); Cleanups: - Remove the last references to EXPORT_OP_ASYNC_LOCK - Make file_remove_privs_flags() static - Remove redundant __GFP_NOWARN when GFP_NOWAIT is used - Use try_cmpxchg() in start_dir_add() - Use try_cmpxchg() in sb_init_done_wq() - Replace offsetof() with struct_size() in ioctl_file_dedupe_range() - Remove vfs_ioctl() export - Replace rwlock() with spinlock in epoll code as rwlock causes priority inversion on preempt rt kernels - Make ns_entries in fs/proc/namespaces const - Use a switch() statement() in init_special_inode() just like we do in may_open() - Use struct_size() in dir_add() in the initramfs code - Use str_plural() in rd_load_image() - Replace strcpy() with strscpy() in find_link() - Rename generic_delete_inode() to inode_just_drop() and generic_drop_inode() to inode_generic_drop() - Remove unused arguments from fcntl_{g,s}et_rw_hint() Fixes: - Document @name parameter for name_contains_dotdot() helper - Fix spelling mistake - Always return zero from replace_fd() instead of the file descriptor number - Limit the size for copy_file_range() in compat mode to prevent a signed overflow - Fix debugfs mount options not being applied - Verify the inode mode when loading it from disk in minixfs - Verify the inode mode when loading it from disk in cramfs - Don't trigger automounts with RESOLVE_NO_XDEV If openat2() was called with RESOLVE_NO_XDEV it didn't traverse through automounts, but could still trigger them - Add FL_RECLAIM flag to show_fl_flags() macro so it appears in tracepoints - Fix unused variable warning in rd_load_image() on s390 - Make INITRAMFS_PRESERVE_MTIME depend on BLK_DEV_INITRD - Use ns_capable_noaudit() when determining net sysctl permissions - Don't call path_put() under namespace semaphore in listmount() and statmount()" * tag 'vfs-6.18-rc1.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (38 commits) fcntl: trim arguments listmount: don't call path_put() under namespace semaphore statmount: don't call path_put() under namespace semaphore pid: use ns_capable_noaudit() when determining net sysctl permissions fs: rename generic_delete_inode() and generic_drop_inode() init: INITRAMFS_PRESERVE_MTIME should depend on BLK_DEV_INITRD initramfs: Replace strcpy() with strscpy() in find_link() initrd: Use str_plural() in rd_load_image() initramfs: Use struct_size() helper to improve dir_add() initrd: Fix unused variable warning in rd_load_image() on s390 fs: use the switch statement in init_special_inode() fs/proc/namespaces: make ns_entries const filelock: add FL_RECLAIM to show_fl_flags() macro eventpoll: Replace rwlock with spinlock selftests/proc: add tests for new pidns APIs procfs: add "pidns" mount option pidns: move is-ancestor logic to helper openat2: don't trigger automounts with RESOLVE_NO_XDEV namei: move cross-device check to __traverse_mounts namei: remove LOOKUP_NO_XDEV check from handle_mounts ...